Molecular community data meets anaerobic digestion Model 1 (ADM1) - a study about the correlation between...

Advanced models, such as the Anaerobic Digestion Model No 1 (ADM1), are essential for operating, planning, and optimizing renewable energy production in anaerobic digester plants (AD-P)s. In this study, the ADM1da model was employed to simulate a two-step AD-P in an industrial setting. The ADM1da model is an extended ADM1 model for mixed substrates, accounting for substrate-specific disintegration, temperature effects, biogas-related mass reduction, and mineral solids content. ADM1 models can represent the anaerobic digestion processes, although the biological assumptions are coarse and reflect the knowledge and available tools for microbial communities at the time of development. Meanwhile, metagenome-centric metaproteomics provides deeper insight into the metabolic activities of microbial communities in AD-Ps. Until now, this data has not been integrated with ADM1 models. The objective of this study is to assess the feasibility of incorporating metagenome-centric metaproteome data into the ADM1 model. In a novel approach, 49 high-quality metagenome-assembled genomes (MAGs) with associated protein abundances were systematically classified into the trophic groups defined by the ADM1 model using specifically developed grouping rules. Abundances of MAGs were more variable than the process parameter-dependent dynamics of ADM1. Depending on the grouping rules, 32%-78% of all high-quality MAGs were successfully categorized into ADM1 trophic groups. However, some MAGs, e.g., Methanotrix, were multifunctional (acetoclastic and hydrogenotrophic methanogenesis) and required assignment to multiple groups. Unfortunately, more precise grouping rules resulted in greater discrepancies between metaproteomics data and the model. Additionally, 22% of the MAGs could not be assigned. The metagenome-centric metaproteome data imply that ADM1 probably needs extension to cover the observed microbial function of syntrophic acetate oxidizers, hydrolytic bacteria, lactate- and ethanol-fermenting bacteria, and mortality by phages. It was also observed that changes in process parameters, such as those caused by seasonal feeding, led to significant changes in the protein abundance Integrating metagenome-centric metaproteomic data into ADM1 trophic groups was shown to be feasible.Some trophic groups detected in protein data but not implemented in ADM1 imply the need for data-driven model enhancement and approval. In the future, more accurate models considering molecular data could support a deeper understanding of microbial community dynamics in AD-Ps.